Rooibos Is Caffeine-Free. It Was Never Decaffeinated.

Rooibos has never held a milligram of caffeine, not before harvest, not after fermentation, not in the cup. That is a different fact from "decaffeinated," which means caffeine was there and was taken out, usually leaving a trace behind. Rooibos skips that whole process because the plant never had anything to remove.

The plant makes no caffeine, and never has

Aspalathus linearis is a legume, in the same family as peas, lupins, and clover, and further still from the true tea plant, Camellia sinensis, than most drinkers assume. A 1989 survey of the genus Aspalathus by South African botanists Van Wyk and Verdoorn tested the whole group for alkaloids, the class of compounds caffeine belongs to, and found almost none. A trace of an unrelated alkaloid called sparteine turned up in a few species, but no caffeine, no relative of caffeine, nothing from that branch of plant chemistry at all. The rooibos bush was not stripped of caffeine. It was never built with the machinery to make it.

That "machinery" has a name. Caffeine is built in three chemical steps, each one a methylation, meaning a small carbon-and-hydrogen group gets bolted onto the molecule in sequence. Each step needs its own enzyme, and each of those enzymes is coded for by a gene. Camellia sinensis has those genes. Coffee has its own, independently evolved version of the same genes. Cacao has a third, separate version again. A 2016 study in the Proceedings of the National Academy of Sciences traced this out precisely: caffeine has evolved at least five separate times across unrelated plant lineages, tea, coffee, cacao, citrus, and guarana, each one recruiting old enzymes that originally did something else entirely and repurposing them, through a handful of mutations, into a caffeine-making toolkit. It is a textbook case of convergent evolution, the same trick invented from scratch again and again because it works.

Rooibos's family tree simply never ran that experiment. The Crotalarieae, the legume tribe rooibos belongs to, shows no sign of ever recruiting those enzymes for that job. So there is no half-finished pathway lying dormant in the plant's genes, no caffeine gene switched off. The biochemical starting point coffee, tea, and cacao all separately built on is simply absent from rooibos.

Why "caffeine-free" and "decaffeinated" are not the same claim

Decaffeinated coffee and decaffeinated black tea both started life with caffeine already in the leaf or bean, the way every other product of their own family does, and it was removed afterward by one of a small handful of industrial processes. The figure most widely cited for how much has to come out, in the United States, is 97 percent of the original caffeine, a bar high enough to matter but never claiming completeness. The remaining few percent is why a cup of decaf coffee still carries something, typically a handful of milligrams, rather than a true zero.

There are a few ways the caffeine actually comes out. The oldest solvent methods rinse the beans or leaves with a chemical, historically benzene, now more often ethyl acetate or, less commonly, methylene chloride, which binds to the caffeine and is then washed away. The newer supercritical carbon dioxide method instead pressurizes CO2 into a state that behaves like both a gas and a liquid, circulates it through the beans, and lets the caffeine dissolve into the CO2 while the flavor compounds, being larger molecules, stay largely put. A third option, the Swiss Water Process, uses no chemical solvent at all: beans sit in hot water, the water is filtered through activated carbon to strip out caffeine while sparing the larger flavor molecules, and that flavor-saturated water is reused to soak the next batch, drawing out its caffeine by nothing more than the water no longer having a caffeine gradient left to draw from.

Every one of those methods is a subtraction. Something had to be there to take away, and something is inevitably left. Rooibos has no such story to tell, because there was never an addition to begin with.

What that actually means in the cup

Put a number on it and the difference sharpens. An 8-ounce cup of black tea generally carries somewhere in the range of 40 to 70 milligrams of caffeine; green tea usually runs lower, roughly 20 to 45 milligrams; a similar cup of drip coffee is higher still, close to 95 milligrams. Decaffeinated versions of any of those do not reach zero. A decaffeinated 8-ounce cup typically still carries a few milligrams, commonly cited in the range of 2 to 5, the residue that 97 percent removal leaves behind. Rooibos sits under none of those figures, because it was never above zero to begin with. A cup, or a pot left steeping for forty minutes, or a second and third refill from the same leaves, carries the same true zero throughout.

This is also why rooibos never turns the bitter, over-extracted way a black tea can. Bitterness in true tea comes largely from tannins releasing as the leaf steeps too long, a separate chemical story from caffeine, but the two facts travel together in the reader's mind because both are questions about what a "safe, gentle" cup can and cannot do. Rooibos is low in tannin as well as free of caffeine, and neither fact required anything to be processed out. Both were simply how the leaf came.

The honest edge of the claim

None of this means every trace compound in rooibos has been hunted down and ruled out by every laboratory that has ever looked, and a careful reader deserves that qualifier stated plainly rather than glossed over. What the botanical survey work shows is an absence of the alkaloid class caffeine belongs to across the genus, not a guarantee that no laboratory anywhere could ever detect a vanishingly small trace under sufficiently sensitive instruments. That is a normal scientific caveat, the kind that applies to almost any "zero" claim in biology, and it does not weaken the practical answer. For a plant with no caffeine-synthesis genes and no history of producing the compound, "caffeine-free" is the accurate description, not a marketing rounding-off of "very low."

The distinction is worth holding onto past this page. When a label says decaffeinated, ask what was removed and how much is left. When it says caffeine-free, as rooibos rightly is, there is a shorter question behind it: whether the plant ever had the chemistry to make the stuff at all. Rooibos never did.

Sources

1. Aspalathus linearis, SANBI PlantZAfrica, on the plant's legume family placement and caffeine-free status.
2. Alkaloids of the genera Aspalathus, Rafnia and Wiborgia (Fabaceae-Crotalarieae), Van Wyk and Verdoorn, South African Journal of Botany, 1989, on the genus-wide alkaloid survey finding no caffeine and only trace sparteine.
3. Convergent evolution of caffeine in plants by co-option of exapted ancestral enzymes, Huang et al., PNAS, 2016, on the at-least-five independent evolutionary origins of caffeine biosynthesis.
4. The coffee genome provides insight into the convergent evolution of caffeine biosynthesis, Denoeud et al., Science, 2014, on coffee's independently evolved caffeine-synthesis genes.
5. Caffeine in coffee: its removal. Why and how?, Ramalakshmi and Raghavan, Critical Reviews in Food Science and Nutrition, 1999, on decaffeination by organic solvents, water, and supercritical CO2.
6. Decaf coffee, National Coffee Association, on the roughly 97 percent caffeine-removal figure for decaffeinated coffee.
7. Molecular Cloning and Functional Characterization of Three Distinct N-Methyltransferases Involved in the Caffeine Biosynthetic Pathway in Coffee Plants, Ogawa et al., Plant Physiology, 2001, on the three sequential methylation enzymes that build caffeine.